Transcript assembly of RNA-seq data from different tissues and individuals

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

Transcript assembly of RNA-seq data from different tissues and individuals

Quanwei Zhang
Hello:

I am trying to assemble transcripts using RNA-seq data by the tool Trinity, which will be used for gene annotation for Maker. Now I have data from two tissues with two replicates each. Should I merge all four samples to get one assembly file? Or should I merge replicates of each tissue separately and use the two assembly files as input of Maker. Merging all samples into one, we will have much higher coverage level, but I think there may be some genes expressed by tissue-specific isoforms. So I not sure whether I should merge RNA-seq from different tissues.
What's more, I find some published RNA-seq data from another individual (and also for different tissue from us) for the same species. Should I merge all RNA-seq together (across individuals and tissues)? Or should I generate different transcript assembly and use all those assemblies as input to Maker?

Thanks
Best
Quanwei

_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: Transcript assembly of RNA-seq data from different tissues and individuals

Michael Campbell
I would probably try merging the replicates but not the tissues. You can then pass the output files to MAKER in a comma separated list in the opts file.

Example:
est=/PATH/TO/file1.fsata,/PATH/TO/file2.fasta

Good luck,
Mike

> On Jan 31, 2017, at 2:08 PM, Quanwei Zhang <[hidden email]> wrote:
>
> Hello:
>
> I am trying to assemble transcripts using RNA-seq data by the tool Trinity, which will be used for gene annotation for Maker. Now I have data from two tissues with two replicates each. Should I merge all four samples to get one assembly file? Or should I merge replicates of each tissue separately and use the two assembly files as input of Maker. Merging all samples into one, we will have much higher coverage level, but I think there may be some genes expressed by tissue-specific isoforms. So I not sure whether I should merge RNA-seq from different tissues.
> What's more, I find some published RNA-seq data from another individual (and also for different tissue from us) for the same species. Should I merge all RNA-seq together (across individuals and tissues)? Or should I generate different transcript assembly and use all those assemblies as input to Maker?
>
> Thanks
> Best
> Quanwei
> _______________________________________________
> maker-devel mailing list
> [hidden email]
> http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org


_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: Transcript assembly of RNA-seq data from different tissues and individuals

Fields, Christopher J
I agree with Mike.  I also suggest not combining RNA-Seqs from different runs (e.g. different studies) even if they are from the same tissue, development stage etc. There are many other factors (biological variation, sample quality, sequencing chemistry or technology differences, etc) that can significantly and negatively impact trx assembly quality.

chris

On 1/31/17, 1:26 PM, "maker-devel on behalf of Michael Campbell" <[hidden email] on behalf of [hidden email]> wrote:

    I would probably try merging the replicates but not the tissues. You can then pass the output files to MAKER in a comma separated list in the opts file.
   
    Example:
    est=/PATH/TO/file1.fsata,/PATH/TO/file2.fasta
   
    Good luck,
    Mike
   
    > On Jan 31, 2017, at 2:08 PM, Quanwei Zhang <[hidden email]> wrote:
    >
    > Hello:
    >
    > I am trying to assemble transcripts using RNA-seq data by the tool Trinity, which will be used for gene annotation for Maker. Now I have data from two tissues with two replicates each. Should I merge all four samples to get one assembly file? Or should I merge replicates of each tissue separately and use the two assembly files as input of Maker. Merging all samples into one, we will have much higher coverage level, but I think there may be some genes expressed by tissue-specific isoforms. So I not sure whether I should merge RNA-seq from different tissues.
    > What's more, I find some published RNA-seq data from another individual (and also for different tissue from us) for the same species. Should I merge all RNA-seq together (across individuals and tissues)? Or should I generate different transcript assembly and use all those assemblies as input to Maker?
    >
    > Thanks
    > Best
    > Quanwei
    > _______________________________________________
    > maker-devel mailing list
    > [hidden email]
    > http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
   
   
    _______________________________________________
    maker-devel mailing list
    [hidden email]
    http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
   

_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: Transcript assembly of RNA-seq data from different tissues and individuals

Quanwei Zhang
Thank you guys for your suggestions. So you do not suggest to use RNA-seq data from another study, even I assemble them separately and then provide both assemblies into Maker as a comma separated list. The issues you mentioned do exist, but some people did collect RNA-seq data from different individuals and used them for gene annotation (e.g., doi:10.1038/ng.3198). But thank you for your suggestions, I will think about it.

Best
Quanwei

2017-01-31 16:05 GMT-05:00 Fields, Christopher J <[hidden email]>:
I agree with Mike.  I also suggest not combining RNA-Seqs from different runs (e.g. different studies) even if they are from the same tissue, development stage etc. There are many other factors (biological variation, sample quality, sequencing chemistry or technology differences, etc) that can significantly and negatively impact trx assembly quality.

chris

On 1/31/17, 1:26 PM, "maker-devel on behalf of Michael Campbell" <[hidden email] on behalf of [hidden email]> wrote:

    I would probably try merging the replicates but not the tissues. You can then pass the output files to MAKER in a comma separated list in the opts file.

    Example:
    est=/PATH/TO/file1.fsata,/PATH/TO/file2.fasta

    Good luck,
    Mike

    > On Jan 31, 2017, at 2:08 PM, Quanwei Zhang <[hidden email]> wrote:
    >
    > Hello:
    >
    > I am trying to assemble transcripts using RNA-seq data by the tool Trinity, which will be used for gene annotation for Maker. Now I have data from two tissues with two replicates each. Should I merge all four samples to get one assembly file? Or should I merge replicates of each tissue separately and use the two assembly files as input of Maker. Merging all samples into one, we will have much higher coverage level, but I think there may be some genes expressed by tissue-specific isoforms. So I not sure whether I should merge RNA-seq from different tissues.
    > What's more, I find some published RNA-seq data from another individual (and also for different tissue from us) for the same species. Should I merge all RNA-seq together (across individuals and tissues)? Or should I generate different transcript assembly and use all those assemblies as input to Maker?
    >
    > Thanks
    > Best
    > Quanwei
    > _______________________________________________
    > maker-devel mailing list
    > [hidden email]
    > http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org


    _______________________________________________
    maker-devel mailing list
    [hidden email]
    http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org




_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: Transcript assembly of RNA-seq data from different tissues and individuals

Carson Holt-2
I think he means not to combine them for the transcript assembly preparation (i.e. assembly them separately). But you still provide them all to maker as a comma separated list.

—Carson
 
On Jan 31, 2017, at 2:33 PM, Quanwei Zhang <[hidden email]> wrote:

Thank you guys for your suggestions. So you do not suggest to use RNA-seq data from another study, even I assemble them separately and then provide both assemblies into Maker as a comma separated list. The issues you mentioned do exist, but some people did collect RNA-seq data from different individuals and used them for gene annotation (e.g., doi:10.1038/ng.3198). But thank you for your suggestions, I will think about it.

Best
Quanwei

2017-01-31 16:05 GMT-05:00 Fields, Christopher J <[hidden email]>:
I agree with Mike.  I also suggest not combining RNA-Seqs from different runs (e.g. different studies) even if they are from the same tissue, development stage etc. There are many other factors (biological variation, sample quality, sequencing chemistry or technology differences, etc) that can significantly and negatively impact trx assembly quality.

chris

On 1/31/17, 1:26 PM, "maker-devel on behalf of Michael Campbell" <[hidden email] on behalf of [hidden email]> wrote:

    I would probably try merging the replicates but not the tissues. You can then pass the output files to MAKER in a comma separated list in the opts file.

    Example:
    est=/PATH/TO/file1.fsata,/PATH/TO/file2.fasta

    Good luck,
    Mike

    > On Jan 31, 2017, at 2:08 PM, Quanwei Zhang <[hidden email]> wrote:
    >
    > Hello:
    >
    > I am trying to assemble transcripts using RNA-seq data by the tool Trinity, which will be used for gene annotation for Maker. Now I have data from two tissues with two replicates each. Should I merge all four samples to get one assembly file? Or should I merge replicates of each tissue separately and use the two assembly files as input of Maker. Merging all samples into one, we will have much higher coverage level, but I think there may be some genes expressed by tissue-specific isoforms. So I not sure whether I should merge RNA-seq from different tissues.
    > What's more, I find some published RNA-seq data from another individual (and also for different tissue from us) for the same species. Should I merge all RNA-seq together (across individuals and tissues)? Or should I generate different transcript assembly and use all those assemblies as input to Maker?
    >
    > Thanks
    > Best
    > Quanwei
    > _______________________________________________
    > maker-devel mailing list
    > [hidden email]
    > http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org


    _______________________________________________
    maker-devel mailing list
    [hidden email]
    http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org



_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org


_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org
Reply | Threaded
Open this post in threaded view
|

Re: Transcript assembly of RNA-seq data from different tissues and individuals

Fields, Christopher J

Exactly

 

chris

 

From: Carson Holt <[hidden email]>
Date: Tuesday, January 31, 2017 at 3:35 PM
To: Quanwei Zhang <[hidden email]>
Cc: Chris Fields <[hidden email]>, "[hidden email]" <[hidden email]>
Subject: Re: [maker-devel] Transcript assembly of RNA-seq data from different tissues and individuals

 

I think he means not to combine them for the transcript assembly preparation (i.e. assembly them separately). But you still provide them all to maker as a comma separated list.

 

—Carson

 

On Jan 31, 2017, at 2:33 PM, Quanwei Zhang <[hidden email]> wrote:

 

Thank you guys for your suggestions. So you do not suggest to use RNA-seq data from another study, even I assemble them separately and then provide both assemblies into Maker as a comma separated list. The issues you mentioned do exist, but some people did collect RNA-seq data from different individuals and used them for gene annotation (e.g., doi:10.1038/ng.3198). But thank you for your suggestions, I will think about it.

Best

Quanwei

 

2017-01-31 16:05 GMT-05:00 Fields, Christopher J <[hidden email]>:

I agree with Mike.  I also suggest not combining RNA-Seqs from different runs (e.g. different studies) even if they are from the same tissue, development stage etc. There are many other factors (biological variation, sample quality, sequencing chemistry or technology differences, etc) that can significantly and negatively impact trx assembly quality.

chris


On 1/31/17, 1:26 PM, "maker-devel on behalf of Michael Campbell" <[hidden email] on behalf of [hidden email]> wrote:

    I would probably try merging the replicates but not the tissues. You can then pass the output files to MAKER in a comma separated list in the opts file.

    Example:
    est=/PATH/TO/file1.fsata,/PATH/TO/file2.fasta

    Good luck,
    Mike

    > On Jan 31, 2017, at 2:08 PM, Quanwei Zhang <[hidden email]> wrote:
    >
    > Hello:
    >
    > I am trying to assemble transcripts using RNA-seq data by the tool Trinity, which will be used for gene annotation for Maker. Now I have data from two tissues with two replicates each. Should I merge all four samples to get one assembly file? Or should I merge replicates of each tissue separately and use the two assembly files as input of Maker. Merging all samples into one, we will have much higher coverage level, but I think there may be some genes expressed by tissue-specific isoforms. So I not sure whether I should merge RNA-seq from different tissues.
    > What's more, I find some published RNA-seq data from another individual (and also for different tissue from us) for the same species. Should I merge all RNA-seq together (across individuals and tissues)? Or should I generate different transcript assembly and use all those assemblies as input to Maker?
    >
    > Thanks
    > Best
    > Quanwei
    > _______________________________________________
    > maker-devel mailing list
    > [hidden email]
    > http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org


    _______________________________________________
    maker-devel mailing list
    [hidden email]
    http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org

 

_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org

 


_______________________________________________
maker-devel mailing list
[hidden email]
http://box290.bluehost.com/mailman/listinfo/maker-devel_yandell-lab.org